The current proposal is guided by the premise that there are two principal functions of motion processing. One is to make perceptual information accessible to cognitive processing by specifying a unique 3D object that is perceived as meaningful, i.e., a representation that makes contact with stored knowledge. The other function is to extract information from the transforming optical array for specifying the layout of surfaces and guiding self-movements. In general, the perceptual information necessary for representation is object-centered, whereas the information necessary for specifying self-motion is viewer-centered. It is hypothesized that changes in sensitivity to each of these two sources of information are a function of different development factors. The specific objectives for this funding period are to further evaluate and refine the preceding proposal. Three related issues in the development of motion processing will be investigated by testing infants from 2 to 15 months of age. The first is concerned with the spatial and temporal limits on perceiving motion. A series of experiments are proposed to examine developmental changes in threshold sensitivity to spatial and temporal displacements in random dot kinematograms. Adult analogue experiments using spatially filtered displays will be used to provide convergent evidence. The second issue is concerned with the development of object-centered processes and involves experiments designed to examine infants' responses to moving displays. It is our contention that these processes undergo a developmental shift between 3 and 5 months of age. Whereas, 3-month-old infants respond to constituent properties (e.g. local rigidity), 5-month- old infants respond to the changes in the perception of self- motion. Experiments are proposed to evaluate how changes in the range of exploratory behaviors (e.g., looking locomotion) demand new sensitivities to optical flow for guiding responses. A wide range of paradigms and measures will be used in this research. The procedures includes habituation, forced-preferential looking, and a moving room analogue involving projection screens; the measures include looking time, heart rate, facial expression, foot switches, and postural sway. This diversity of methods and questions will provide new insights into those perceptual processes that are fundamental to perceptual competence as well as those that are dynamically driven by the specific developmental tasks encountered by infants.